Google DeepMind CEO Warns AI Could Repeat Social Media's Divisive, Addictive Patterns
Demis Hassabis, CEO of Google DeepMind, has raised alarm about the potential dangers of artificial intelligence, warning that it could repeat the divisive and addictive patterns seen in social media. He criticized Silicon Valley's 'move fast and break things' mentality, urging companies to consider long-term societal consequences.
Hassabis pointed out that social media platforms, such as X, Instagram, and Facebook, have fostered addiction and division among users. Frequent use can hijack the brain's dopamine pathways, leading to dependency similar to substance addiction. This is exacerbated by engagement traps that cause inflammatory content to spread faster, as it triggers strong emotions. Users tend to form homophilic clusters, creating echo chambers that amplify confirmation bias.
Hassabis warned that AI risks repeating these toxic patterns. He criticized the 'move fast and break things' mentality, which prioritizes rushing products to market without adequately considering long-term societal consequences. He warned that AI could create the same issues as social media, but on a much larger scale. With AGI estimated to be around five to ten years away, Hassabis urged for caution and responsible development.
Hassabis' warnings highlight the need for careful consideration of AI's potential impacts. As we approach AGI, it's crucial to learn from the mistakes of social media and prioritize the long-term benefit of society over short-term gains. Studies show that spending more than two hours a day on social media can reduce prefrontal impulse control by 35%, underscoring the importance of responsible use and development.
